Client

Energy and Renewables

Industry

Energy and Power Generation

Primary goal

Process optimisation, asset management

Technologies

Machine learning (multi-layer perceptrons), real-time analytics, IoT data integration

Digital twin solutions can be used for real-time monitoring of solar power plant blocks, enhancing asset insights and asset management.

The client was looking for a solution to forecast inverter efficiency using machine learning models to enable proactive measures.

KPMG in India created a 3D digital twin of the solar power plant for real-time visualisation of the farm to improve operational efficiency.

The challenge

The ability to predict power generation ahead of schedule is critical in solar power generation, as significant penalties are incurred if the asset fails to meet committed output in future time blocks to the grid.

Maintaining inverter availability and health is a key asset management requirement for solar power generation fleets.

The opportunity

Analysed historical data (past 1 year) to understand solar generation and irradiance trends
Cleaned and stored data on a cloud platform (Data Explorer)
Developed dashboards for anomaly detection and visualisation (Plotly curves) using Azure Data Explorer
Created digital twin models for each solar asset, integrated with real-time data

Integrated external irradiance APIs with real-time plant data to predict DC power output and inverter downtime
Developed predictive models using ML techniques (Decision Trees, Random Forest, KNN, etc.)
Built ensemble models (5–6 models) to generate weighted predictions and reduce error
Trained and tested models on 6 months of data and evaluated using MAE, MSE, and MAPE

KPMG in India’s approach included:

Deep domain understanding of power generation
Data integration across LIMS, OSI PI, and cloud platforms
Stack build with OPC UA, API layers, and time-series databases

Advanced modelling for predictive accuracy and asset management
Real-time deployment with operational dashboards
Action enablement through SOP-driven interventions

The impact

solar_power

Digital Twin of client Solar Power Plant Block improved the operational performance largely by increasing their MTBF by 6 per cent and reducing MTTR 8 per cent by predicting faults ahead of time

currency_rupee

Solution lowered the revenue losses occurring due to penalties incurred in failing to provide committed power output in the future time blocks to the Grid due to variable weather conditions by 9 per cent

How we make the difference

KPMG in India leverages advanced analytics and real-time data integration to enable intelligent asset management in industrial environments. Our solutions help organisations optimise performance, improve quality, and achieve operational excellence through AI-driven insights.